Research
Working Papers
Learning Traps
July 2023
Under policy uncertainty, policy choices generate a tradeoff between payoffs and information. Incumbent leaders who set policies can implement their preferred policy, but if the outcome is bad, constituents learn this and threaten office removal. I study a model of how this threat of information revelation constrains leaders’ policies. I show that, in equilibrium, leaders implement uninformative “learning trap” policies that halt learning, even when more preferable policies are available. I argue that the codification of the commune in Imperial Russia can be viewed as a learning trap that distorted peasants’ economic incentives to obfuscate the benefits of liberal reform.
A New Interpretation of Productivity Growth Dynamics in the Pre-Pandemic and Pandemic-Era U.S. Economy, 1950-2023 (with Robert J. Gordon)
August 2023, NBER Working Paper 30267
This paper provides a unified framework that explains variations of productivity growth in both the pre-pandemic and pandemic-era U.S. economy. It resolves the puzzles of why productivity growth was so rapid during the recessions of 2008-09 and 2020 yet so slow in 2010-19 and 2021-23, as well as why productivity growth was procyclical in 1950-85 and 2010-19 but less so in 1986-2006. Our approach treats productivity growth as a residual, equal by definition to output growth minus growth in hours of work. The key insight is that in response to the collapse of output in the 2008-09 and 2020 recessions, business firms overreacted with “excess layoffs,” adjusting hours to the output decline with a much higher elasticity than normal. Our regression analysis, which allows post-recession rehiring that gradually unwinds the excess layoffs, explains why productivity growth was so high in 2009 and why it was so low in 2010-19 as rehiring boosted hours growth. Post-sample simulations explain why productivity growth was so high in 2020 and why it fell to a negative -1.1 percent in the nine quarters of 2021-23. The paper concludes by suggesting implications for the future long-term evolution of productivity growth in the business sector and total economy. A new quarterly productivity data file for 17 industries provides additional perspectives. In 2009 the excess decline in hours and surge in productivity growth were entirely attributable to the three industries most impacted by the recession -- manufacturing, construction, and finance. In 2020-23 the turnaround from strongly positive to negative productivity growth primarily occurred in the goods sector. In contrast 2020-23 productivity growth was uniformly strong in work-from-home service industries. We relate the work-from-home productivity surge to recent survey evidence reporting self-assessments that workers at home perceive their own productivity as having improved.
Download here. Press coverage in The Economist, NBER Digest.
Work in Progress
When Do Governors Experiment?
Large scale experiments often generate flow utility effects; for example, voters benefit from participating in policy experiments that are likely to help them. How do the flow utility effects of experiments affect term-limited principals’ incentives to acquire information? Under two-term limits, I show that leaders acquire no information if an experiment will likely fail and acquire maximal information if an experiment will surely succeed. Under sufficient uncertainty, leaders “gamble for resurrection” — gathering less information as a policy becomes more likely to succeed. While increasing the fierceness of (electoral) competition during the experimentation process increases the amount of information leaders gather, increasing competition after information is acquired decreases the amount of information gathered. Extending term limits to three periods may encourage more efficient information gathering.
I augment the insights of the model using new data on US state governors’ willingness to propose new policies. I construct a panel dataset of State of the State speeches from 1996-2020 for all 50 US states. I fine-tune a set of Roberta models to identify snippets of text corresponding to “policies," “policy proposals," and "novel policies." I demonstrate a negative correlation between gubernatorial approval and frequency of novel policy proposals policies in non-election years, followed by a flat or positive correlation in election years. These results are driven primarily by tax cuts and internal government reform from low approval governors; as well as crime policy and healthcare reform.
Inverted Digital Property Rights
Consumers often learn about the value of innovative products via other consumers' feedback. A larger quantity of an item sold hence generates more information about the value of an innovation. Consequently, traditional patent policy — where innovators are awarded with monopoly patent rights to recoup fixed costs — may not be optimal, as their underprovision of goods generates underprovision of information. Instead, it may be optimal for regulators to invert property rights: allow competition, where comeptitiors can imitate an innovator's product, during a first trial period; endow an innovator with monopoly rights in a second period; and then permit competition again in a third period. I apply these insights to the algorithmic design of content production and incentivization on platforms like TikTok.
Learning Under Grievance Procedures (with Kim Sarnoff)
(Description forthcoming)
Model Based Inference (with Kim Sarnoff)
This is an experimental project that uses a "coefficient weight" method to elicit how participants' create mental models that organize data; and how those mental models vary and update with the structure of the data generating process.
Publications (Macroeconomics)
Transatlantic Technologies: The Role of ICT in the Evolution of U.S. and European Productivity Growth (with Robert J. Gordon)
Spring 2020, International Productivity Monitor No. 38
We examine the role of the ICT revolution in driving productivity growth behavior for the United States and an aggregate of ten Western European nations (the EU-10) from 1977 to 2015. We find that the standard growth accounting approach is deficient when it separates sources of growth between ICT capital deepening and TFP growth, because much of the effect of the ICT revolution was channeled through spillovers to TFP growth rather than being limited to the capital deepening path- way. Using industry-level data from EU KLEMS, we find that most of the 1995-2005 U.S. productivity growth revival was driven by ICT-intensive industries producing market services and computer hardware. In contrast the EU-10 experienced a 1995- 2005 growth slowdown due to a paucity of ICT investment, a failure to capture the efficiency benefits of ICT, and performance shortfalls in specific industries in- cluding ICT production, finance-insurance, retail-wholesale, and agriculture. After 2005 both the United States and the EU-10 suffered a growth slowdown, indicating that the benefits of the ICT revolution were temporary rather than providing a new permanent era of faster productivity growth. Also circulated as NBER WP 27425.
Download here. Press coverage in NBER Digest, VoxEU.
The Industry Anatomy of the Transatlantic Productivity Growth Slowdown: Europe Chasing the American Frontier (with Robert J. Gordon)
Fall 2019, International Productivity Monitor No. 37
By merging KLEMS data covering 16 industry groups within the total economy and 11 manufacturing sub-industries, we compare and contrast productivity growth from 1950 to 2015 in the United States with an aggregate of the ten largest European nations (EU-10) from 1972 to 2015. We interpret the EU-10 performance as catching up to the United States in stages. Strikingly, the total economy "early-to-late" productivity growth slowdown from 1972-1995 to 2005-2015 in the EU-10 (-1.68 percentage points) was almost identical to the U.S. slowdown from 1950-1972 to 2005-2015 (-1.67 percentage points). There is a very high EU-U.S. correlation in the magnitude of the early-to-late slowdown in each industry, suggesting that the productivity growth slowdown from the early postwar years to the most recent decade was due to a retardation in technical change that affected the same industries by roughly the same magnitudes on both sides of the Atlantic. Also circulated as NBER WP 25703.
Download here. Press coverage in The Economist, VoxEU.